As a learning designer, I’ve been exploring how AI can improve my work. Every time I dive into a tutorial or test a shiny new feature, I end up overwhelmed by jargon, edge cases, and hype. Upskilling for AI experience design can be frustrating, leaving me with one key question: What makes some AI skills stick while others fade with the next platform update?
Staying current in the age of agentic AI is no longer about mastering the latest interface. Tools change too quickly to keep up. What matters more is understanding how AI systems work: how they think, process, and evolve. Shifting from features to frameworks moves you from trend follower to forward builder by focusing more on the architecture, data flows, and design logic behind them than the specific tools.?
Here’s what we’ll cover:
Design how you learn and remember
Build fluency with core AI concepts
Look past features and focus on capabilities
Pay attention to the ecosystem, not the trends
Shift to systems thinking?
Upskilling for AI experiences
Design how you learn and remember
Take learning design as an example. A beginner might follow a template to build a course. But a seasoned designer understands:
- how people learn,
- how memory works,
- and how feedback supports behavior change.
That knowledge helps them to adapt to new formats and avoid shallow engagement. The same mindset applies to AI. When you understand the system, you’re not just reacting to change. You’re ready to shape what comes next.
This is the heart of upskilling for AI experiences: learning systems, not just surfaces. This shift is not abstract. It’s deeply practical. So how do we build that mindset? Let’s look at what it means in practice.
Build fluency with core AI concepts
Core AI concepts are more than buzzwords. They represent underlying capabilities – such as function calling and retrieval augmented generation – that apply across platforms. Understanding concepts can help you shape outcomes, not just follow steps.
For designers, it’s crucial to understand how knowledge is structured and how language is processed. This includes knowing the difference between taxonomy and ontology, structured and unstructured data, prompts versus function calls, and large language models and large action models. This knowledge is no longer just for engineers; it’s part of the modern design and learning toolkit.
How to start learning:
- Focus on one concept at a time. Select a core AI concept and break it into manageable chunks to support active engagement.
- Reinforce through repetition. Use flashcards, diagrams, or quick summaries to strengthen memory and recall over time.
- Apply in context. Connect the concept to a real example. Prototype something small or trace how the idea shows up in a Salesforce tool.
Look past features and focus on capabilities
It’s easy to get distracted by how a tool looks or its demo features. But remember, surface features are temporary. The underlying capability – such as summarization, classification, search, or translation – is what endures even when brands or UIs change.
By focusing on the capability, you can more easily compare tools, envision new uses, and adapt your strategy as tools evolve. Understanding the capability means you can apply it broadly across tools.
A good example is reasoning and decision support. Whether you’re in customer service, onboarding, or product design, systems that reason through context and suggest next steps will increasingly appear. If you understand how systems use knowledge graphs or disambiguate multiple meanings, you can design experiences that support it, measure its value, and plan for its evolution.
How to start learning:
- Connect concepts to real tools. Start with core capabilities like summarization or classification. Look for how they show up across different tools.
- Support memory with structure. Organize what you learn in a visual map or list. Seeing patterns across tools helps reinforce recall.
- Apply and reflect. Test a capability in a real project or Salesforce feature. Use feedback or self-reflection to turn insight into lasting skill.
Pay attention to the ecosystem, not the trends
AI moves fast, and the news cycle is full of product releases, trend reports, predictions, and corporate statements. But to forecast where things are actually headed, you need to look beyond the headlines.
Big shifts often start in how things are built and only later show up in branding. Paying attention to the ecosystem gives you a longer runway to spot patterns before they hit the mainstream – and it gives you a major advantage.?
For example, the emergence of Model Component Protocol (MCP) might not make every keynote, but it has implications for how tools interconnect. Being aware of it early means you can design with interoperability in mind.
API design is another area to watch. Today’s powerful capabilities, like smart retrieval or cross-system orchestration, depend on how systems communicate. Reading API documentation and understanding how these connections work gives you insight into what’s possible, even if you’re not writing the code.
How to start learning:
- Scan trusted sources regularly. Set aside time each week to scan developer forums, technical blogs, or standard-setting efforts.
- Explore one concept deeply. Pick an emerging concept, like MCP, and study how it could impact interoperability or system design.
- Connect changes to design. As you learn, consider how shifts in APIs or data models affect your work. Apply that insight to plan more flexible, future-ready experiences.
Shift to systems thinking
Digital fluency today is about understanding the intelligence behind the interfaces. Beyond creating screens or documents, we’re shaping systems that respond, adapt, and make decisions. This means thinking in terms of inputs and outputs, signals and intent, feedback loops, and logic models.
It also means designing modular content that can be tagged, indexed, reused, and recombined to build experiences that adapt and proliferate in real time.?
How to start learning:
- Ask systems questions. Shift your focus from tools to architecture. Ask: How does this tool understand inputs? Where does its logic live? How is knowledge stored and applied?
- Map a familiar system. Sketch a simple system you use often. Identify its inputs, outputs, and feedback loops to understand how it adapts to change.
- Trace data flow. Follow how data moves through a real project. Pinpoint where decisions are made and what triggers them to sharpen your systems thinking and design adaptability.
Upskilling for AI Experiences
What finally helped me break through the noise was changing how I thought about learning. I stopped chasing the latest tools and started focusing on the systems underneath. That shift from surface to structure made everything click.
When you understand how AI reasons, stores knowledge, and responds to inputs, you gain more than technical skill. You build durable fluency that helps you design for change rather than react to it. You move from chasing updates to creating value that lasts.
So start where you are. Trace one system. Learn one protocol. Apply one idea. What you learn today will shape what’s possible tomorrow.
